$$ \footnotesize Dev Patel \\ [email protected] \\ devpatelio.github.io \\

$$

Abstract

The future of programmable, precise, and personalized therapeutics can be realized through mRNA therapies, yet the biggest bottleneck right now in increasing the effectiveness and breadth of available therapies comes down to the structure of the mRNA molecule and how this structure ultimately dictates the immunogenicity, RNA-protein binding affinity, and the possible diversity of functions these therapies can have. By leveraging novel machine learning models [specifically using reinforcement learning with GNN policy networks], generating optimal secondary structure arrangements and tertiary folds for mRNA caps can allow us to increase our immune repertoire while producing novel and targeted therapies for key therapeutic areas [programmable, precise, and personalized]. While there are different other ways to tackle some of the foundational problems with the effectiveness of mRNA therapies, interpreting the structure and relative function of mRNA molecules offers new insights on designing novel platforms to rapidly test and screen possible candidates, ultimately making mRNA the de facto modality for several emerging therapeutic areas.

Looking for help with anyone who has worked with rich datasets across different domains + specifically understands how to model and interact different biological processes into a common baseline for knowledge. Specifically labs @PennState or @Alphafold and @Stanford Researchers who've worked in functional molecular biology to some extent. Perhaps those who have access to running experiments in the wetlab to provide insight on how folding works in the real world + experts at companies leading in making mRNA therapeutics to learn more about how computational results translate into real products.

1. Introduction

• Current Status and Future Potential ****

      Key Questions: 
  

  1. For mRNA molecules, how much can we actually engineer onto the molecule in terms of the length to the structure and even the complexity of folds so that they can effectively be translated into proteins? [how does energy constraints play into this and are there any specific points to note about this?]

  2. I'm guessing that most of our mRNA therapies are based on specific antibody-coding sequences, but how does personalization specifically benefit certain diseases like cancer, and could there be a "general" solution if we were able to sequence several thousand cancer patients?

2. Fundamentals of Pharmacology

• Funamdentlas of mRNA Pharmocology

3. Status Quo Pipeline

• Manufacturing Challenges and Bottlenecks

      Key Questions: 
  

  1. How does the nanoparticle coating, or other traditional manufacturing procedures, impact

  2. I'm guessing that most of our mRNA therapies are based on specific antibody-coding sequences, but how does personalization specifically benefit certain diseases like cancer, and could there be a "general" solution if we were able to sequence several thousand cancer patients?

4. Challenges in Delivery

• Opportunities and Challenges in Delivery

5. Possible Pipeline for Approach

• mRNA Optimization for Design Pipeline

6. RNA Design for Folding to Secondary Structure